Abstract
Herbal teas and other herbal preparations are becoming more and more popular, and
it is essential to ensure their quality. Quality control methods that are simple,
fast, and of low cost are needed by the producers and by inspections. Infrared spectroscopy
coupled with multivariate mathematical methods has been shown to be useful for the
identification and characterization of plant samples. In this work, we developed a
method for the identification of herbal drugs in different herbal teas. 100 one-component
herbal teas were first used to build an identification algorithm,
which showed 100 % correct classification. In the next validation step, 13 samples
from 7 herbal mixtures were analyzed, confirming high accurate results for classification.
The influence of using different number of components in the principal component analysis
is also explored. Infrared spectroscopy coupled with analysis of variance, principal
component analysis, and discriminant analysis was shown to be highly applicable for
quality control procedures.
Key words
herbal teas - mid-infrared spectroscopy - chemometrics - quality control
